hmmm...
I happened to know, that currently, HSPICE doesn't support any phase noise simulation.
Disappointed ?
But, I also happended to know, that labled as 2004.3 and planned to be release on someday of 2004, a RF version HSPICE will be out.
This version would be good at dealing with phase noise, mixer, S-parameter, stability, G_u, ...
Now it is in beta testing procedure and is calling for testing customer. Maybe you could contact with them.
The phase noise can be infered from HSpice by running the FFT command. Then the phase noise can be calculated by taking the difference between the peak of your signal to its sideband.